1 See e.g. Wilson, R. McL., Studies in the Gospel of Thomas (London: Mowbray, 1960)Google Scholar 49; Gärtner, B. E., The Theology of the Gospel of Thomas (London: Collins, 1961)Google Scholar 67; Haenchen, E., ‘Literatur zum Thomasevangelium’ (Part I), ThR 27 (1961)Google Scholar 147–78 (175); Cullmann, O., ‘The Gospel of Thomas and the Problem of the Age of the Tradition Contained Therein: A Survey’, Interpretation 16 (1962) 418–37CrossRefGoogle Scholar (436).

2 Tuckett, C. M., ‘Thomas and the Synoptics’, NovT 30 (1988) 132–57Google Scholar.

3 For example, Tuckett, ‘Thomas and the Synoptics’, refers to GTh 5 and 31 (145–6, 143), but not to the other passages discussed below: his article is not so narrowly focused on Luke as the present study; rather it claims to deal with ‘some examples’ (145) across the whole Synoptic tradition.

4 Riley, G. J., ‘The Influence of Thomas Christianity on Luke 12:14 and 5:39’, HTR 88 (1995) 229–34CrossRefGoogle Scholar.

5 Riley, ‘Influence of Thomas Christianity’, 229.

6 Johnson, S. R., Seeking the Imperishable Treasure: Wealth, Wisdom, and a Jesus Saying (Eugene, OR: Cascade, 2008) esp. 58–79Google Scholar.

7 See e.g. (among many other places), the clear statement in Quispel, G., ‘The Gospel of Thomas and the New Testament’, VC 11 (1957) 189–207Google Scholar (194); also Quispel, , ‘L'Évangile selon Thomas et les Clémentines’, VC 12 (1958) 181–96Google Scholar; Quispel, , ‘Some Remarks on the Gospel of Thomas’, NTS 5 (1958–59) 276–90CrossRefGoogle Scholar; Quispel, , ‘L'Évangile selon Thomas et le Diatessaron’, VC 13 (1959) 87–117Google Scholar.

8 J. Sieber, ‘A Redactional Analysis of the Synoptic Gospels with regard to the Question of the Sources of the Gospel according to Thomas’ (PhD diss., Claremont University, 1965).

9 Schramm, T., Der Markus-Stoff bei Lukas: Eine literarkritische und redaktionsgeschichtliche Untersuchung (Cambridge: Cambridge University, 1971) esp. 10–21CrossRefGoogle Scholar (16, 20–1).

10 Patterson, S. J., The Gospel of Thomas and Jesus (Sonoma, CA: Polebridge, 1993)Google Scholar.

11 DeConick, A. D., The Original Gospel of Thomas in Translation: With a Commentary and New English Translation of the Complete Gospel (London/New York: Continuum, 2006)Google Scholar.

12 DeConick, Original, 23.

13 Gregory, A. F., The Reception of Luke and Acts in the Period before Irenaeus: Looking for Luke in the Second Century (WUNT 2/169; Tübingen: Mohr, 2003)Google Scholar 353.

14 DeConick, Original, 23. The phrase appears in italics in the original.

15 Schürmann, H., ‘Das Thomasevangelium und das lukanische Sondergut’, BZ 7 (1963) 236–60Google Scholar.

16 Dehandschutter, B., ‘L'Évangile selon Thomas: Témoin d'une tradition prélucanienne?’, L'Évangile selon Luc: Problèmes littéraires et théologiques. Memorial Lucien Cerfaux (ed. Neirynck, F.; BETL 32; Gembloux: Duculot, 1973) 287–97Google Scholar, 324–6.

17 Bovon, F., ‘Les sentences propres à Luc dans l'Évangile selon Thomas’, Colloque internationale: ‘L'Évangile selon Thomas et les Textes de Nag Hammadi’, Québec, 29–31 Mai 2003 (ed. Painchaud, L. and Poirier, P.-H.; Louvain: Peeters, 2007) 43–58Google Scholar; now also in English translation: ‘Sayings Specific to Luke in the Gospel of Thomas’, New Testament and Christian Apocrypha: Collected Studies (Tübingen: Mohr, 2009) 161–73Google Scholar.

18 Bovon, ‘Les sentences’, 47.

19 Bovon, ‘Les sentences’, 49–50.

20 Bovon, ‘Les sentences’, 53.

21 Bovon, ‘Les sentences’, 55.

22 Gregory, Reception, 14; DeConick, Original, 190.

23 Tuckett, ‘Thomas and the Synoptics’, 142.

24 Tuckett, ‘Thomas and the Synoptics’, 146 and n. 49.

25 See the stress on unity/division in GTh 4, 22, 23, 48, 61, 106, and perhaps 11, 47, 89, 108.

26 Currently, M. Goodacre is probably the most prominent Q sceptic. See especially his The Case Against Q: Studies in Markan Priority and the Synoptic Problem (Harrisburg, PA: Trinity Press International, 2002)Google Scholar. In addition, one could note the various contributors to Goodacre, M. and Perrin, N., eds., Questioning Q: A Multidimensional Critique (London: SPCK, 2004)Google Scholar, and now Watson, F. B., with his ‘Q as Hypothesis: A Study in Methodology’, NTS 55 (2009) 397–415CrossRefGoogle Scholar. There are important precursors to these in the work of A. Farrer, M. Goulder and E. P. Sanders.

27 Koester, H., Synoptische Überlieferung bei den apostolischen Vätern (TU 65; Berlin: Akademie-Verlag, 1957)Google Scholar.

28 Tuckett, ‘Thomas and the Synoptics’; cf. also Tuckett, , ‘Sources and Methods’, The Cambridge Companion to Jesus (ed. Bockmuehl, M.; Cambridge: Cambridge University, 2001) 121–37CrossRefGoogle Scholar (129).

29 Gregory, Reception, 155–6.

30 Although n.b. ‘the undoubted presence’ of LkR in Thomas (Gregory, Reception, 150).

31 See e.g. Davies, S. and Johnson, K., in their argument that Thomas is the earliest of the Gospels: ‘Mark's use of the Gospel of Thomas’, Parts 1 & 2, Neot 30 (1996) 307–34; and 31 (1997) 233–61Google Scholar; DeConick has argued that the core of Thomas was composed prior to 50 ce (Original Gospel of Thomas, 8). See also Riley and Johnson (to be discussed below), whose view of Thomas's influence upon Luke clearly implies the priority of Thomas. Koester, H. comments that it ‘may well date from the first century’: Robinson, J. M., ed., The Nag Hammadi Library (Leiden: Brill, 3rd ed. 1993)Google Scholar 125.

32 See e.g. DeConick's strong resistance to a Gnostic characterisation, favouring instead that of ‘early Syrian religiosity’ (Original Gospel of Thomas, 4). The most recent substantial discussion of the question, by Marjanen, also gives a negative answer. See Marjanen, A., ‘Is Thomas a Gnostic Gospel?’, Thomas at the Crossroads (ed. Uro, R.; Edinburgh: T. & T. Clark, 1998) 107–39Google Scholar.

33 The present study assumes the substantial unity of Thomas, rather than a very long development over the course of nearly a century, as argued by DeConick. This does not necessarily rule out some instability, although the substantial similarity between the Greek fragments and the Coptic version suggests that forms of the text like our extant texts were those which circulated in antiquity. The similarity in the texts consists in (a) substantial similarity in order, with one exception (Gk GTh 30 = Co GTh 30 + 77.2-3); (b) substantial similarity in extent of sayings, with one exception (Co GTh 36 is an abbreviation of its Greek counterpart). That the Coptic translation was made from a Greek text closely resembling those Greek texts which have survived is evident from (c) the fact that where the Coptic text and the extant Greek manuscripts overlap, the Greek loan words in the Coptic text almost all have correspondingly similar Greek words in the Greek fragments. Of the 27 cases, the only exceptions are a case of ⲁⲗⲗⲁ (←καί in GTh 3.3), an unparalleled use of ⲏ in GTh 30.2, and GTh 32's preference in Coptic for ⲟⲩⲇⲉ over the Greek's οὔτɛ … οὔτɛ …. In the latter two cases the discrepancy arises from a different syntax in the surrounding context, and the variation between ⲟⲩⲇⲉ and οὔτɛ is insignificant when one considers that Coptic frequently does not distinguish between ⲇ and ⲧ. If one leaves the ⲏ and ⲟⲩⲇⲉ out of account, then one is left with only one exception out of 25 (ⲁⲗⲗⲁ/καί), and particles are the elements least predictably rendered in other Greek-to-Coptic translations.

34 E.g. Pagels, E., Beyond Belief: The Secret Gospel of Thomas (New York: Random House, 2003)Google Scholar 47, and more forcefully Bauckham, R. J., Jesus and the Eyewitnesses (Grand Rapids: Eerdmans, 2006) 236–7Google Scholar.

35 Uro, R., Thomas: Seeking the Historical Context of the Gospel of Thomas (New York: T&T Clark, 2003)Google Scholar 88: ‘Only in Matthew and Thomas does Jesus’ response contain a reference to the divine source of the confession (cf. the blessing in Matt 16:17 and Thomas' intoxication in Gos. Thom. 13:5) which is affirmed with the unique role that Jesus assigns to the disciple who has given the appropriate answer.'

36 DeConick, Original, 129.

37 Finnegan, R., Oral Poetry: Its Nature, Significance and Social Context (Cambridge: Cambridge University, 1977)Google Scholar; Thomas, R., Literacy and Orality in Ancient Greece (Key Themes in Ancient History; Cambridge: Cambridge University, 1992)CrossRefGoogle Scholar.

38 The phrase appears in Finnegan, Oral Poetry, 258.

39 Cf. Snodgrass, K. R., ‘The Gospel of Thomas: A Secondary Gospel’, SecCent 7 (1989–90) 19–38Google Scholar (28); Uro, Seeking the Historical Context, 88–9; Uro, ‘Thomas and Oral Gospel Tradition’, Thomas at the Crossroads (ed. Uro) 8–32 (10); Wood, J. H., ‘The New Testament Gospels and the Gospel of Thomas: A New Direction’, NTS 51 (2005) 579–95CrossRefGoogle Scholar (589). The term is seemingly first applied to gospels scholarship in Kelber, W., The Oral and Written Gospel: The Hermeneutics of Speaking and Writing in the Synoptic Tradition, Mark, Paul, and Q (Philadelphia: Fortress, 1983)Google Scholar 197.

40 This is not to suggest the phases are merely sequential, with the first finishing when the second begins.

41 The reason it is perhaps an inappropriate phrase is that when it was originally coined by Walter Ong in 1971, it applied not to the relationship between two pieces of literature but rather referred to a whole cultural mentality: premodern ‘primary orality’ in contrast with modern ‘secondary orality’. Ong, W. J., Rhetoric, Romance and Technology: Studies in the Interaction of Expression and Culture (Ithaca/London: Cornell University, 1971)Google Scholar 20. Between these two epochs came the interposition of ‘the individualised introversion of the age of writing, print, and rationalism’ (285, where Ong also refers to his belief that he coined the phrases). See further Ong, 's Orality and Literacy: The Technologizing of the Word (London/New York: Methuen, 1982) 136–8CrossRefGoogle Scholar and passim. Although not strictly a pre-modern vs. modern contrast, Ong gives the informative illustration by way of a contrast between the hour-long speeches of the presidential debate between Lincoln and Douglas and the ‘domesticated’ contemporary debates (Orality and Literacy, 137). Secondary orality is a very wide cultural phenomenon ‘with which we are going to have to live through the foreseeable future’ (Ong, Rhetoric, 303).

42 Uro, ‘Thomas and Oral Gospel Tradition’, 31.

43 A good example of attention to textual variants is DeConick's commentary (Original, passim), which appends very useful references to ‘Agreements in Syrian Gospels, Western Text and Diatessaron’ in the discussion of each saying. See also earlier e.g. Quispel, G., ‘L'Evangile selon Thomas et le “Texte Occidental” du Nouveau Testament’, VC 14 (1960) 204–15Google Scholar.

44 It is possible in theory that Lukan influence comes in not (only) (a) at the stage of Thomas's composition, but also (b) in the course of its transmission in Greek, (c) at the point of translation, or (d) in its Coptic transmission. We will note in the conclusion, however, that there is good evidence for influence (a) and (b).

45 Tuckett, ‘Thomas and the Synoptics’, 157.

46 In the interests of full disclosure, it should be noted that this article is based on the assumption of Markan priority but assumes neither Q nor the Farrer hypothesis.

47 Here and below, words of interest for the comparison of the different versions of the sayings in Thomas and the Synoptics are underlined.

48 E.g. Tuckett, ‘Sources and Methods’, 129: ‘This seems to be clear evidence that, at this point at least, Thomas presupposes Luke's finished Gospel.’

49 E.g. … ὃ οὐ ϕανɛ[ρωθήσɛται], which is what Clement has in Strom. 1.13.3.

50 DeConick, Original, 61.

51 As such, pace Gregory, Reception, 155, it is not merely a matter of the shared word δɛκτός.

52 So e.g. Bruce, F. F., ‘The Gospel of Thomas’, Jesus and Christian Origins Outside the New Testament (London: Hodder & Stoughton, 1974) 110–56Google Scholar (127). The second part of the saying in Thomas, ‘no doctor heals those who know him’, is a very peculiar proverb, contradicted by the almost universal practice of doctors in antiquity. Prof. Sir Geoffrey Lloyd has remarked to me as follows: ‘Very curious. No parallels for that remark about doctors not treating those who know them come to mind, and plenty of texts that contradict the principle’ (email communication, 28/02/2008). The combination of the sayings in Thomas may, however, be indebted to the pairing of Luke 4.23-24 or Mark 6.4-5. Patterson, Gospel of Thomas and Jesus, 31–2, followed by Johnson, Seeking the Imperishable Treasure, 77 n. 50, argue that the Thomasine pairing is more original. This is little more than form-critical guesswork, however: assuming that Mark adopts a ‘softening tendency’, replacing the harsh GTh 31.2. Johnson's additional argument for the priority of Thomas is particularly difficult to accept: ‘Note that Luke has Jesus himself stating that this is a common proverb and therefore probably not an original saying of Jesus.’

53 DeConick, Original, 141.

54 There are no compelling reasons for questioning the priority of the Markan version of this parable.

55 The reading ‘a good man’ (xⲣⲏ[ⲥⲧⲟ]ⲥ, rather than xⲣⲏ[ⲥⲧⲏ]ⲥ) is also possible.

56 This should perhaps be emended to: ‘Perhaps they did not recognise him.’

57 Isaiah 5 also surfaces in Mark 12.9 and parallels, but Thomas has ended the parable by this time.

58 Snodgrass, K., The Parable of the Wicked Tenants (Tübingen: Mohr, 1983)Google Scholar 52.

59 Hubaut, M., La parabole des vignerons homicides (Paris: Gabalda, 1976)Google Scholar 134; Snodgrass, Parable, 52–3.

60 DeConick, Original, 215, though she allows for the possibility of secondary orality.

61 Kloppenborg, J., The Tenants in the Vineyard: Ideology, Economics, and Agrarian Conflict in Jewish Palestine (Tübingen: Mohr, 2006)Google Scholar.

62 An example of each can be mentioned. (1) The idea that Mark's ϕραγμός (‘palisade’, ‘wall’, ‘fence’) is a specifically Egyptian viticultural item derived from the lxx (Kloppenborg, Tenants, 168, 172) is puzzling: ‘The reference to the building of a palisade (καὶ πɛριέθηκɛν ϕραγμόν) reflects a specifically Septuagintal addition to the mt and mirrors the Egyptian viticultural practice that had influenced the lxx translators’ (168). But m. Kil. 4.2 discusses the boundary-fence, and the גדר. The biblical occurrences of this Hebrew word are translated in the lxx most frequently by ϕραγμός: moreover, Kloppenborg's references to various Greek writers' uses of this noun (Tenants, 159 n. 25) prove that it is by no means specifically Egyptian. So both the item (the fence) and the terminology for it (ϕραγμός) are clearly unproblematic in a Palestinian context. (2) On the legal side, Kloppenborg argues (Tenants, 330–4) that Thomas's reference to the killing of the heir better reflects law in contrast to Mark 12.7's apparently ludicrous claim that the tenants would inherit. But the reasoning of Thomas's tenants is just as ludicrous: the heir is not the owner of the vineyard, so why should the tenants maintain their ownership by killing the heir? Moreover, as noted above, Thomas gives less of an explanation than do the Synoptics.

63 Kloppenborg, Tenants, 269–76.

64 Patterson, Gospel of Thomas, 51.

65 Kloppenborg, Tenants, 258–9.

66 Kloppenborg, Tenants, 259.

67 Kloppenborg, Tenants, 269.

68 If one were heavily committed to Occam's razor, one could appeal to the text-form found in Luke 11.33 אBCD et al. as sufficient to account for Thomas's phraseology. This would perhaps impose too narrow constraints upon the sources available to Thomas.

69 Quispel, ‘The Gospel of Thomas and the New Testament’, 190–1, thinks that the version in G. Eb. is related to Thomas via the Gospel of the Hebrews, but the similarities between Thomas and G. Eb. here are unremarkable.

70 Patterson, Gospel of Thomas and Jesus, 80–1, 92–3 (reasons on p. 81); Hedrick, C. W., ‘An Anecdotal Argument for the Independence of the Gospel of Thomas from the Synoptic Gospels’, For the Children Perfect Instruction: Studies in Honor of Hans-Martin Schenke (ed. Bethge, H.-G. et al. ; NHMS 54; Leiden: Brill, 2002) 113–26Google Scholar (118–19).

71 On a more minor note, Thomas's version also contains an abbreviation of the Markan version similar to that of Luke.

72 Arai, S., ‘Caesar's, God's and Mine: Mark 12:17 par. and Gos. Thom. 100’, Gnosisforschung und Religionsgeschichte. FS Kurt Rudolph zum 65. Geburtstag (ed. Preissler, H. and Seiwert, H.; Marburg: Diagonal, 1994) 43–8Google Scholar (46).

73 Matt 22.19 also has the imperative ἐπιδɛίξατɛ, however.

74 On a very minor note indeed, Matthew, Luke and Thomas (but not Mark) have a pronominal reference to Jesus in the final apophthegm; only Mark has ὁ δὲ Ἰησοῦς.

75 DeConick, Original, 129.

76 Riley, ‘Influence of Thomas Christianity’, 233.

77 Riley, ‘Influence of Thomas Christianity’, 234.

78 Rather, following the theme of the first half of GTh 47, it seems that the over-riding concern is the incompatibility of opposites. GTh 47.1-2 note the impossibility of riding two horses, drawing two bows and serving two masters. Similarly, GTh 47.3-4 simply refer to the incongruity of an old patch on a new garment, and of new wine in old skins and vice versa; the old is not valued over the new in these cases. It is quite possible that GTh 47.5 values the old wine over the new, but only if one already knows that old wine is preferable.

79 Riley, ‘Influence of Thomas Christianity’, 234.

80 E.g. Caird, G. B., Saint Luke (Pelican New Testament Commentaries; Harmondsworth: Penguin, 1963)Google Scholar 98; Schürmann, H., Das Lukasevangelium (Herders Theologischer Kommentar 3; Freiburg: Herder, 1969)Google Scholar 1.300; Marshall, I. H., Commentary on Luke (NIGTC; Grand Rapids: Eerdmans, 1978)Google Scholar 228; Fitzmyer, J. A., The Gospel of Luke I–IX (AB 28; New York: Doubleday, 1982)Google Scholar 597; Nolland, J., Luke 1–9.20 (WBC 35A; Waco: Word, 1989)Google Scholar 250; Schweizer, E., Das Evangelium nach Lukas (NTD 3; Göttingen: Vandenhoeck & Ruprecht, 3rd ed. 1993) 73–4CrossRefGoogle Scholar.

81 DeConick again emphasises the process of oral transmission (Original, 175), but this need not be pitted against literary influence.

82 Quispel, G., ‘The Discussion of Judaic Christianity’, VC 22 (1968) 81–93Google Scholar (85–6); Baarda, T., ‘Luke 12:13-14: Text and Transmission from Marcion to Augustine’, Early Transmission of the Words of Jesus: Thomas, Tatian and the Text of the New Testament (Amsterdam: Free University, 1983) 117–72Google Scholar; repr. from Judaism, Christianity and Other Greco-Roman Cults: Studies for Morton Smith at Sixty, vol. 1 (Leiden: Brill, 1975) 107–62Google Scholar.

83 Similarly DeConick, Original, 230, taking account of a parallel in an Islamic text.

84 Riley, ‘Influence of Thomas Christianity’, 230.

85 Riley, ‘Influence of Thomas Christianity’, 231.

86 Riley, ‘Influence of Thomas Christianity’, 232.

87 Riley, ‘Influence of Thomas Christianity’, 230.

88 LSJ Suppl., 98, citing ‘Bull. épigr. 1955. 163 (p.57)’.

89 Pippidi, D. M., Inscriptiones Scythiae Minoris Graecae et Latinae. Volumen Primum: Inscriptiones Histriae et Viciniae (Bucharest: Typis Academiae Scientiarum Dacoromanae, 1983)Google Scholar.

90 IHistriae 6, ll. 3–5: τὸ δὲ ἀνάλωμα δοῦναι τὸν οἰκονόμον, μɛρίσαι δὲ τοὺς μɛριστάς; cf. IHistriae 19, ll. 3–5: … [μɛρὶσα]ι δ[ὲ τ]οὺς μɛρ[ιστάς]. Cf. the fully restored instances in IHistriae 21, l. 5 and IHistriae 40, l. 2.

91 See Kern, O., Die Inschriften von Magnesia am Maeander (Berlin: W. Spemann, 1900)Google Scholar 45, and xxx-xxxi for the date.

92 For more on the μɛρισταί, see Bowsky, M. W. Baldwin, ‘Epigrams to the Elder Statesman and a Young Noble from Lato Pros Kamara (Crete)’, Hesperia 58 (1989) 115–29CrossRefGoogle Scholar (122); Henry, A. S., ‘Provisions for the Payment of Athenian Decrees: A Study in Formulaic Language’, ZPE 78 (1989) 247–95Google Scholar for references to the οἱ μɛριζόμɛνοι and the annual μɛρισμός in Athens (261, 263), and further references to the verb μɛρίσαι in contexts similar to those of the Istria and Magnesia inscriptions (268–9, 273–92).

93 Ludwich, A., ‘Ueber die Homerischen Glossen Apions’, Philologus 74 (1917) 205–47CrossRefGoogle Scholar (228, ll. 22–23).

94 Dor. 2.33.4. Pingree, D., ed., Dorothei Sidonii Carmen Astrologicum (Leipzig: Teubner, 1976) 359–60Google Scholar.

95 LSJ, 1104a.

96 Riley (‘Influence of Thomas Christianity’, 230–2) does not state whether he thinks that the actual term μɛριστής was the word used in Greek Thomas. On the basis of his emphasis on Luke's apparent coinage of the word, he might think some other wording was used; on the other hand, Riley may be speaking rhetorically of the situation for the analysis of the Lukan language when one leaves Thomas out of consideration.

97 Johnson, Seeking the Imperishable Treasure, esp. 58–79.

98 Johnson, Seeking the Imperishable Treasure, 77, n. 50; cf. 12.

99 Johnson, Seeking the Imperishable Treasure, 76.

100 Johnson, Seeking the Imperishable Treasure, 63.

101 There are, however, a number of possible equivalents, including ἀπολλύναι (539b).

102 Johnson, Seeking the Imperishable Treasure, 72–3.

103 Johnson, Seeking the Imperishable Treasure, 69.

104 Johnson, in his inquiries as to why Thomas would use ‘Matthew's order of adversities … but Luke's verbs’ (Seeking the Imperishable Treasure, 70), etc. is too insistent upon theological reasons for small differences among versions. Such variation might easily be the result of the vagaries of oral transmission. Johnson, however, operates with a highly scribalised model of dependence, in which every variation needs to be justified.

105 Johnson, Seeking the Imperishable Treasure, 71.

106 Johnson's ‘μὴ τὸν ἀπολλύμɛνον’ is similar to John's language, but only because Johnson's retroversion has made it so. As noted, Thomas's ⲉⲙⲁϥⲱϫ is just as close to Luke's ἀνέκλɛιπτον; on the other hand, Thomas's ⲉϥⲙⲏⲛ ⲉⲃⲟⲗ is a standard equivalent for a participle of μένω, as in Johnson's retroversion, and in John 6.27. The perishing/enduring contrast is conventional, however (e.g. Eccl 7.15's righteous ἀπολλύμɛνος and wicked μένων).

107 Gärtner, Theology, 67.

108 Gärtner, Theology, 66.

109 Tuckett, ‘Thomas and the Synoptics’, 157.

110 I avoid here the debate over the original language of Thomas, taking it to be Greek.